Hi! I just pulled the latest changes from origin/main (commit 4d3d177) and ran into a test failure when running the quality checks. Spent a bit debugging this and found an interesting config mismatch.

Description

The test tests/test_conversation_memory.py::TestConversationMemory::test_add_turn_max_limit fails when .env contains MAX_CONVERSATION_TURNS=40 (which comes from .env.example).

Root Cause

Turns out there's a mismatch between the code default (50 in utils/conversation_memory.py:122) and what .env.example sets (40). The test uses the runtime value from .env, so it creates 40 turns and expects add_turn() to reject adding one more—but the check len(context.turns) >= MAX_CONVERSATION_TURNS (40 >= 40) should catch this, yet it doesn't seem to.

Failure Details

FAILED tests/test_conversation_memory.py::TestConversationMemory::test_add_turn_max_limit - assert True is False

Steps to Reproduce

1. Make sure .env has MAX_CONVERSATION_TURNS=40 (default from .env.example)
2. Run: python -m pytest tests/test_conversation_memory.py::TestConversationMemory::test_add_turn_max_limit -v
3. Test fails

Workaround

Setting MAX_CONVERSATION_TURNS=50 in .env makes the test pass.

Suggested Fix

A few options to fix this:

1. Update .env.example to MAX_CONVERSATION_TURNS=50 (align with code default)
2. Or update the code default to 40 (align with .env.example)
3. Alternatively, make the test explicitly set MAX_CONVERSATION_TURNS=50 via @patch.dict

Environment

• Python: 3.12.11
• pytest: 9.0.1
• Git commit: 4d3d177 (latest from origin/main)